compact-guide

A command that advises whether to use context compaction, which shortens the conversation while keeping its most relevant information.

In plain words
What is it for?
Use it when changing tasks, finishing research or debugging, or starting implementation after planning. It also explains when compaction could make ongoing work harder.
Why use it?
It helps prevent old research, logs, or finished work from crowding out the details needed for the next task.

Command

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/helloruru/claude-memory-engine/compact-guide
Clone the repo
git clone --depth 1 https://github.com/HelloRuru/claude-memory-engine
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 248 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00000 $0.00248
Opus 5 $0.00000 $0.00124
Sonnet 5 $0.00000 $0.00050
Haiku 4.5 $0.00000 $0.00025

Measured 2d ago against content hash aebc85ca1efc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

compact-guide scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

commands/compact-guide.md · 27 lines

What it actually says

/compact-guide -- Smart Context Compression Guide

Help decide when to use /compact based on the current conversation state.

When to Compact

Situation Compact? Why
Research/exploration done, ready to build Yes Research context is large; the plan is what matters
Planning done, starting to code Yes Plan is already in TodoWrite or a file
Debug session finished Yes Debug logs pollute the next task's context
Switching to a completely different task Yes Clear irrelevant context
Deployment done, starting new feature Yes Deployment logs aren't needed anymore

When NOT to Compact

Situation Why
In the middle of writing code Lose variable names, file paths, partial state
Debugging an unresolved issue Lose error messages and attempted solutions
Multi-file refactor in progress Forget which files still need changes
Just read an important doc, about to reference it Doc content gets summarized away

Auto-Suggest

When context usage exceeds 60%, proactively suggest whether compacting is appropriate based on the current task state.

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 27 lines · 0 tokens per session scan A aebc85ca1efc

Subscribe to this mod's changes

compact-guide is a command published in the GitHub repository HelloRuru/claude-memory-engine (133 stars, last pushed 3mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 248 tokens. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.